Risk & Allocation
Rebalancing is designed to feel wrong
It requires selling what has done well and buying what has done badly, which is exactly what your instincts object to.
This is less a set of instructions about rebalancing a portfolio than an argument, and it is worth saying so at the start.
The argument in brief
- Rebalancing restores intended risk rather than improving returns.
- The discomfort comes from acting against recent performance.
- A written rule removes the need to decide each time.
What it is actually for
Left alone, a portfolio drifts toward whatever has risen most, so its risk level increases without any decision being made. Rebalancing returns the allocation to the level you chose, which is a risk-control action. Whether it also improves returns depends on the period and the assets, and the evidence is mixed rather than conclusive.
Judging it on returns rather than on risk control leads to abandoning it at the wrong moment.
Why it feels wrong
The action is to reduce what has performed well and add to what has performed poorly, which contradicts recency directly. It also feels like interfering with something that is working, which is a strong and misleading intuition. The discomfort is reliable enough that it can be used as a check: if rebalancing feels comfortable, the drift may have been small.
The feeling is not information about markets.
Rules beat judgement
Two common approaches are calendar-based, such as annually, and threshold-based, acting when an allocation drifts beyond a set band. Both work by removing the decision, which matters more than the small differences between them.
Rebalancing more frequently increases costs and taxes without clear benefit, so precision is not the goal. The rule should be written down with the target allocation so the review is a comparison rather than a debate.
Cheaper ways to do it
Directing new contributions toward the underweight allocation rebalances gradually without selling anything. In taxable accounts this avoids realising gains, which can matter substantially depending on local rules. Where withdrawals are being taken, drawing from the overweight allocation does the same job in reverse.
For most people these two mechanisms handle the majority of drift without any explicit trade.
When drift is large
After an extended period, the correction can be big enough that costs and tax become significant considerations. Rebalancing in stages, or over more than one tax period where relevant, is a common practical response. The rules governing this differ substantially by country and by account type.
For a large correction with tax consequences, regulated advice locally is worth the cost.
Some of this will suit you and some will not, and that is the point.
Do not rebalance into a forecast
Rebalancing is not a market call and should not be adjusted because something looks likely to continue rising. Once you start varying the target based on expectations, you have replaced a rule with a judgement. The target itself should change only when your horizon or circumstances change.
On an ordinary week, keeping those two decisions separate is what preserves the discipline.
The takeaway
If rebalancing feels comfortable, check the drift. The discomfort usually means it is working.
Small and repeatable beats ambitious and abandoned, almost every time.
Questions readers ask
How often should I rebalance?
Annually, or when an allocation drifts beyond a set band, is sufficient for most long-horizon portfolios. More frequent rebalancing adds cost without clear benefit.
Does rebalancing increase returns?
Sometimes and not reliably; the evidence varies by period and asset mix. Its dependable function is keeping risk at the level you chose.
